[
https://issues.apache.org/jira/browse/OPENJPA-317?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12520407
]
Craig Russell commented on OPENJPA-317:
---------------------------------------
I like the separation proposed here. Just a few comments.
1. In FetchPlanImpl, you have:
+ public boolean isEnlistInQueryResultCache() {
+ return _fetch.getEnlistInQueryCache();
Why shouldn't both methods have the same signature?
2. Since OpenJPAEntityManagerSPI extends OpenJPAEntityManager, any class that
implements OpenJPAEntityManagerSPI doesn't also need to implement
OpenJPAEntityManager.
3. OpenJPAEntityManagerFactory used to extend Closeable. It might be useful for
it to continue to extend Closeable, as the only method in Closeable is close().
4. I think that these methods should be considered to be part of
OpenJPAEntityManagerFactory and not SPI, since they don't depend on internal
state or internal classes:
+ public void addLifecycleListener(Object listener, Class... classes);
+ public void removeLifecycleListener (Object listener);
> API formalization pre-1.0
> -------------------------
>
> Key: OPENJPA-317
> URL: https://issues.apache.org/jira/browse/OPENJPA-317
> Project: OpenJPA
> Issue Type: New Feature
> Components: jpa
> Affects Versions: 0.9.0, 0.9.6, 0.9.7
> Reporter: Patrick Linskey
> Fix For: 1.0.0
>
> Attachments: OPENJPA-317.patch
>
>
> This issue tracks the effort to formalize and optimize our API prior to the
> 1.0 release.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
